cvmil/swin-base-patch4-window7-224_augmented-v2_tl

swin-base-patch4-window7-224_rice-leaf-disease-augmented-v2_tl

This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6677
• Accuracy: 0.7857

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0003
• train_batch_size: 128
• eval_batch_size: 128
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine_with_restarts
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.9001	1.0	63	1.6795	0.4613
1.429	2.0	126	1.2172	0.6310
1.0492	3.0	189	1.0030	0.6905
0.8729	4.0	252	0.8982	0.7381
0.7743	5.0	315	0.8246	0.7321
0.7084	6.0	378	0.7977	0.7440
0.6631	7.0	441	0.7650	0.7649
0.6279	8.0	504	0.7327	0.7619
0.6004	9.0	567	0.7189	0.7768
0.577	10.0	630	0.7078	0.7798
0.5625	11.0	693	0.6952	0.7738
0.5449	12.0	756	0.6857	0.7857
0.537	13.0	819	0.6802	0.7827
0.5301	14.0	882	0.6746	0.7857
0.5224	15.0	945	0.6715	0.7857
0.5188	16.0	1008	0.6704	0.7857
0.5153	17.0	1071	0.6685	0.7857
0.5112	18.0	1134	0.6676	0.7857
0.5119	19.0	1197	0.6678	0.7857
0.5112	20.0	1260	0.6677	0.7857

Framework versions

• Transformers 4.48.3
• Pytorch 2.5.1+cu124
• Datasets 3.3.2
• Tokenizers 0.21.0